Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

The BQNN Classical Baseline: Constructive Falsification of Near-Term Quantum Advantage at a Fifteen-Million-to-One Energy Disadvantage

Rowan Brad Quni-Gudzinas

Published Jul 27, 2026• 0 citations
Publisher

Abstract

We present a constructive falsification of the quantum advantage claims in Lakhdar-Hamina et al. (2025, PRL). A classical BinaryConnect binarized MLP ensemble achieves 92.73% accuracy on BQNN's 55-image MNIST protocol, exceeding quantum inference (~82-85%) by 7-11pp at 0.0365 J vs 550,000 J (ratio 1.5e7). CAL-BQNN-05 confirmed one year early. arXiv:2507.21222v2 audit series v2.

DOI
10.5281/zenodo.21623218

Related papers

Auditing the BQNN: Does a Tunable Quantum Neural Network on Trapped-Ion and Superconducting Hardware Demonstrate a Route to Near-Term Quantum Advantage?

Rowan Brad Quni-Gudzinas

We audit the BQNN paper by Lakhdar-Hamina et al. (2025, PRL / arXiv:2507.21222v2) which implements a tunable quantum neural network on three quantum computing platforms. Through a complete Phase 1-4 research pipeline including a 5-adversary red-team audit, we evaluate every claim against the Physics of Computation criterion (joules-per-solution < classical alternative). We find the paper's strongest claim (multi-platform QNN benchmarking methodology) is sustained, while its quantum advantage claims are not sustained. Includes 5 calibration register predictions for future verification.

•Jul 25, 2026•0 citations

Evaluating the Advantage 2 Quantum Annealer Prototype: A Comparative Evaluation with Advantage 1 and Hybrid Solver and Classical Restricted Boltzmann Machines on MNIST Classification

Hadi Salloum, Laith Nayal, Manuel Mazzara

Springer proceedings in mathematics & statistics•Jul 16, 2025•0 citations

Bayesian convolutional front-end based uncertainty-aware hybrid quantum–classical image classification

C. W. Wang, Chenhao Huang, Yujing Fan +2 more

Quantum machine learning on noisy intermediate-scale quantum (NISQ) devices often suffers from noise sensitivity, small-data overfitting, and miscalibrated predictive confidence. We propose an uncertainty-aware hybrid Bayesian quantum neural network (BQNN) that couples a Bayesian convolutional front-end with a parameterized quantum circuit (PQC), forming a three-stage pipeline: Bayesian feature extraction, quantum state evolution, and classical decision making. For 28 ×28 grayscale inputs, the front-end uses an 8-filter 3 ×3 Bayesian convolution followed by 2 ×2 max-pooling and a fully connected layer to produce a 10-dimensional embedding, which is mapped to four rotation angles and encoded on 4 qubits. Two parameterized quantum circuit designs are considered for the quantum classification layer. On MNIST and Fashion-MNIST, repeated experiments over five independent random seeds show that the proposed BQNN variants consistently outperform the corresponding QCNN baselines. On MNIST, BQNN-2 achieves a mean best test accuracy of 95.46 ± 0.36%, a mean final test accuracy of 95.44 ± 0.38%, and a mean final test loss of 0.169 ± 0.012. On Fashion-MNIST, BQNN-1 attains the best mean overall performance, with a mean best test accuracy of 87.59 ± 0.62%, a mean final test accuracy of 87.49 ± 0.73%, and a mean final test loss of 0.385 ± 0.017, while BQNN-2 exhibits slightly smaller run-to-run variation. Additional representative comparisons with QNN baselines under the same training configuration further suggest faster convergence and higher peak accuracy for BQNN. Using 30 Monte-Carlo forward passes, BQNN achieves better calibration (ECE 0.007-0.009) than QCNN (ECE 0.016-0.020), thereby reducing overconfident predictions. Additional robustness experiments under noisy mixed-state simulation further show that, under depolarizing noise, BQNN consistently preserves higher peak test accuracy than QCNN across both circuit variants on MNIST, demonstrating improved robustness in NISQ-like noisy settings.

Scientific Reports•Jun 1, 2026•0 citations

A Path Towards Obtaining Quantum Advantage in Training Classical Deep Generative Models with Quantum Priors

Walter Vinci, Lorenzo Buffoni, H. Sadeghi +3 more

Bulletin of the American Physical Society•Mar 6, 2020•0 citations

Quantum Extreme Reservoir Computing Against Matched, Tuned Classical Feature Maps: A Static and Temporal Audit

Nidhi Rajiv Rai, Sandip Tukaram Shingade

Quantum Extreme Reservoir Computing (QERC) has been proposed for machine learning on near-term quantum processors. We audit static and temporal QERC against matched, equally tuned classical baselines, following a protocol committed to version control before test-set evaluation. We reproduce the canonical MNIST baseline (N = 10, D = 1024) with test accuracy 0.9688 (published: 0.970). On compact MNIST (N = 6, n_train = 1000), the disordered quantum reservoir exceeds tuned random Fourier features (C-RFF, within the specified search grid) by +3.87 pp (95% CI [+2.69, +5.12]) and shows no significant difference from tuned tanh features (-0.38 pp, 95% CI [-1.07, +0.58]), while parameter-free cosine encoding, raw PCA features and a trained SVM reach higher mean accuracy. The C-RFF gap depends on its selected scale sitting at the lower grid boundary (on an exploratory extended grid it shrinks to +0.33 pp, 95% CI [-0.62, +1.48]), and on MNIST at N = 10 tuned tanh features are significantly more accurate than the reservoir (delta = -0.94 pp, 95% CI [-1.36, -0.45]). In the canonical regime on MNIST, C-RFF is 0.83 pp more accurate than R-CAN-ZZ: the pre-specified superiority hypothesis for R-CAN-ZZ is not supported, although the difference lies within a +/-1 pp equivalence margin. Static QERC has no delayed memory under an independent-input, reset-per-sample protocol; in the temporal density-matrix formulation and hyperparameter grids studied, a matched echo state network has higher memory capacity (8.457 vs 3.352) and lower NARMA10 error (0.304 vs 0.790) at D = 21.

•Oct 10, 2026•0 citations

No Quantum Advantage Without Structure: A Controlled Benchmark of Quantum Machine Learning on Classical Data

Sadia Afrin Purba, M A Al Mamun, Iyad Obeid A +1 more

•Aug 7, 2026•0 citations
We present a constructive falsification of the quantum advantage claims in Lakhdar-Hamina et al. (2025, PRL). A classical BinaryConnect binarized MLP ensemble achieves 92.73% accuracy on BQNN's 55-image MNIST protocol, exceeding quantum inference (~82-85%) by 7-11pp at 0.0365 J vs 550,000 J (ratio 1.5e7). CAL-BQNN-05 confirmed one year early. arXiv:2507.21222v2 audit series v2.
We audit the BQNN paper by Lakhdar-Hamina et al. (2025, PRL / arXiv:2507.21222v2) which implements a tunable quantum neural network on three quantum computing platforms. Through a complete Phase 1-4 research pipeline including a 5-adversary red-team audit, we evaluate every claim against the Physics of Computation criterion (joules-per-solution < classical alternative). We find the paper's strongest claim (multi-platform QNN benchmarking methodology) is sustained, while its quantum advantage claims are not sustained. Includes 5 calibration register predictions for future verification.
Quantum machine learning on noisy intermediate-scale quantum (NISQ) devices often suffers from noise sensitivity, small-data overfitting, and miscalibrated predictive confidence. We propose an uncertainty-aware hybrid Bayesian quantum neural network (BQNN) that couples a Bayesian convolutional front-end with a parameterized quantum circuit (PQC), forming a three-stage pipeline: Bayesian feature extraction, quantum state evolution, and classical decision making. For 28 ×28 grayscale inputs, the front-end uses an 8-filter 3 ×3 Bayesian convolution followed by 2 ×2 max-pooling and a fully connected layer to produce a 10-dimensional embedding, which is mapped to four rotation angles and encoded on 4 qubits. Two parameterized quantum circuit designs are considered for the quantum classification layer. On MNIST and Fashion-MNIST, repeated experiments over five independent random seeds show that the proposed BQNN variants consistently outperform the corresponding QCNN baselines. On MNIST, BQNN-2 achieves a mean best test accuracy of 95.46 ± 0.36%, a mean final test accuracy of 95.44 ± 0.38%, and a mean final test loss of 0.169 ± 0.012. On Fashion-MNIST, BQNN-1 attains the best mean overall performance, with a mean best test accuracy of 87.59 ± 0.62%, a mean final test accuracy of 87.49 ± 0.73%, and a mean final test loss of 0.385 ± 0.017, while BQNN-2 exhibits slightly smaller run-to-run variation. Additional representative comparisons with QNN baselines under the same training configuration further suggest faster convergence and higher peak accuracy for BQNN. Using 30 Monte-Carlo forward passes, BQNN achieves better calibration (ECE 0.007-0.009) than QCNN (ECE 0.016-0.020), thereby reducing overconfident predictions. Additional robustness experiments under noisy mixed-state simulation further show that, under depolarizing noise, BQNN consistently preserves higher peak test accuracy than QCNN across both circuit variants on MNIST, demonstrating improved robustness in NISQ-like noisy settings.
Quantum Extreme Reservoir Computing (QERC) has been proposed for machine learning on near-term quantum processors. We audit static and temporal QERC against matched, equally tuned classical baselines, following a protocol committed to version control before test-set evaluation. We reproduce the canonical MNIST baseline (N = 10, D = 1024) with test accuracy 0.9688 (published: 0.970). On compact MNIST (N = 6, n_train = 1000), the disordered quantum reservoir exceeds tuned random Fourier features (C-RFF, within the specified search grid) by +3.87 pp (95% CI [+2.69, +5.12]) and shows no significant difference from tuned tanh features (-0.38 pp, 95% CI [-1.07, +0.58]), while parameter-free cosine encoding, raw PCA features and a trained SVM reach higher mean accuracy. The C-RFF gap depends on its selected scale sitting at the lower grid boundary (on an exploratory extended grid it shrinks to +0.33 pp, 95% CI [-0.62, +1.48]), and on MNIST at N = 10 tuned tanh features are significantly more accurate than the reservoir (delta = -0.94 pp, 95% CI [-1.36, -0.45]). In the canonical regime on MNIST, C-RFF is 0.83 pp more accurate than R-CAN-ZZ: the pre-specified superiority hypothesis for R-CAN-ZZ is not supported, although the difference lies within a +/-1 pp equivalence margin. Static QERC has no delayed memory under an independent-input, reset-per-sample protocol; in the temporal density-matrix formulation and hyperparameter grids studied, a matched echo state network has higher memory capacity (8.457 vs 3.352) and lower NARMA10 error (0.304 vs 0.790) at D = 21.